OBJECT RECOGNITION FOR AUGMENTED REALITY APPLICATIONS

Developments in the field of neural networks, deep learning, and increases computing systems’ capacity have allowed for a significant performance boost scene semantic information extraction algorithms their respective mechanisms. The work presented this paper investigates various object classification- recognition frameworks proposes novel framework, which incorporates Super-Resolution as preprocessing method, along with YOLO/Retina network component. resulting analysis framework was fine-tuned benchmarked using COCO dataset, results being encouraging. can potentially be utilized, not only still image scenarios but also video processing.